diff --git a/test_common.py b/test_common.py
index 03c061b..d73e436 100644
--- a/test_common.py
+++ b/test_common.py
@@ -5,8 +5,8 @@ import re
 import sys
 import time
 import fcntl
+import pickle
 import struct
-import serial
 import socket
 import subprocess
 
@@ -35,6 +35,7 @@ class DeviceUnderTest:
 class DeviceUnderTestAeadUARTP(DeviceUnderTest):
     def __init__(self, ser=None):
         self.ser = ser
+        self.firmware_path = None
 
     def prepare(self):
         exp_hello = b"Hello, World!"
@@ -450,20 +451,27 @@ class OpenOcd:
 
 def run_nist_lws_aead_test(vectors_file, build_dir,
                            logic_mask=0xffff,
-                           basecls=DeviceUnderTestAeadUARTP):
+                           basecls=DeviceUnderTestAeadUARTP,
+                           results_path=None):
+    results = {}
+
     kat = list(parse_nist_aead_test_vectors(vectors_file))
+    results['test_vectors'] = kat
 
     dut = basecls(build_dir)
 
+    fw_path = dut.firmware_path
+    if fw_path is not None:
+        with open(fw_path, 'rb') as fw_file:
+            firmware = fw_file.read()
+        results['firmware'] = (fw_path, os.stat(fw_path), firmware)
+
     dut.flash()
     dut.prepare()
     sys.stdout.write("Board prepared\n")
     sys.stdout.flush()
 
     dump_a = dut.dump_ram()
-    time_dips = None
-    ram_utilization = None
-    rom_utilization = None
 
     try:
         tool = LogicMultiplexerTimeMeasurements(logic_mask)
@@ -476,11 +484,9 @@ def run_nist_lws_aead_test(vectors_file, build_dir,
 
             if i == 1 and dump_a is not None:
                 dump_b = dut.dump_ram()
+                results['ram_dumps'] = [dump_a, dump_b]
                 longest = compare_dumps(dump_a, dump_b)
                 print("  longest chunk of untouched memory = %d" % longest)
-                ram_utilization = longest
-
-        time_dips = tool.capture
 
     except Exception as ex:
         print("TEST FAILED")
@@ -489,55 +495,12 @@ def run_nist_lws_aead_test(vectors_file, build_dir,
     finally:
         tool.end_measurement()
 
-    return time_dips, ram_utilization, rom_utilization
-
-
-def main(argv):
-    if len(argv) < 3:
-        print("Usage: test_common.py port LWC_AEAD_KAT.txt")
-
-    eprint(argv[0])
-    script_dir = os.path.split(argv[0])[0]
-    if len(script_dir) > 0:
-        os.chdir(script_dir)
-
-    kat = list(parse_nist_aead_test_vectors(argv[2]))
-
-    dev = argv[1]
-    ser = serial.Serial(dev, baudrate=115200, timeout=5)
-    dut = DeviceUnderTestAeadUARTP(ser)
-
-    try:
-        tool = SaleaeTimeMeasurements()
-        tool.begin_measurement()
-        dut.flash()
-        eprint("Flashed")
-        dut.prepare()
-        eprint("Prepared")
-        sys.stdout.write("Hello, World!\n")
-        sys.stdout.flush()
-
-        dump_a = dut.dump_ram()
-
-        for i, m, ad, k, npub, c in kat:
-            tool.arm()
-            run_nist_aead_test_line(dut, i, m, ad, k, npub, c)
-            tool.unarm()
-
-            if dump_a is not None and i == 1:
-                dump_b = dut.dump_ram()
-                longest = compare_dumps(dump_a, dump_b)
-                print("  longest chunk of untouched memory = %d" % longest)
-
-    except Exception as ex:
-        print("TEST FAILED")
-        raise ex
-
-    finally:
-        tool.end_measurement()
-        sys.stdout.flush()
-        sys.stderr.flush()
+    results['time_dips'] = tool.capture
 
+    if results_path is None:
+        results_path = os.path.join(build_dir, 'results.pickle')
+    if results_path is not False:
+        with open(results_path, 'wb') as f:
+            pickle.dump(results, f, pickle.HIGHEST_PROTOCOL)
 
-if __name__ == "__main__":
-    sys.exit(main(sys.argv))
+    return results